Color identifying of vehicles based on color container and BP network

On the basis of model building and extracting of moving vehicles, the paper propose a new method to identify the vehicle color based on color container and BP neural network. The method analyze the contribution of three channel of the HSI color space, extract the vehicle features using container, design a special neural network for the experiment. The experimental results show that the accuracy rate of the color identifying is high and the error rate is low using this method, which overcome the short-comings of the conventional method.